An explainable deep learning framework based on sparse projected residual networks to predict fluid, crystallized, and total intelligence from resting-state functional magnetic resonance imaging in 5,285 participants from the Adolescent Brain Cognitive Development study suggests that intelligence emerges from the interaction between localized computational hubs and distributed communication pathways.
Abstract
Mapping resting-state brain organization to individual differences in cognitive ability remains a major challenge in population neuroinformatics. Although deep learning enables flexible modeling of brain connectivity, limited interpretability restricts its scientific and clinical utility. To address this objective, we developed an explainable deep learning framework based on sparse projected residual networks to predict fluid, crystallized, and total intelligence from resting-state functional magnetic resonance imaging in 5,285 participants from the Adolescent Brain Cognitive Development study. We incorporated three complementary explainability methods (Integrated Gradients, Gradient Shapley Additive Explanations, and Occlusion) to interpret model behavior. The framework outperformed existing approaches, achieving Pearson correlations of 0.44, 0.58, and 0.56 for fluid, crystallized, and total intelligence, respectively, corresponding to predictive improvements of 6 to 9 percent. All three explainability methods produced near-identical feature rankings (pairwise rank correlations greater than 0.99). Consensus maps revealed a dual-layered functional architecture where primary predictive hubs localized within canonical systems, while the strongest global predictive pathways frequently bypassed these hubs through distributed, long-range relay connections. These findings suggest that intelligence emerges from the interaction between localized computational hubs and distributed communication pathways. Ultimately, these normative network architectures provide clinical reference maps to detect individual deviations, supporting earlier diagnosis, cognitive subtype stratification, and treatment monitoring in atypical neurodevelopment.
Healthy ageing reorganises large-scale functional brain networks. Using resting-state fMRI from 615 adults aged 18 to 88 years in Cam-CAN. We combined raw functional-connectivity analyses, graph-theoretical topology, Ridge and MLP age prediction and their attributions in order to understand how functional brain organis...
Quentin Sénant, Clément Guichet, Nicolas Grivel et al.· bioRxiv· 0 citations
It is demonstrated that the proposed FC-CNN approach successfully decodes brain states from MEG functional connectivity and is promising for discovery of predictive biomarkers for brain disorders.
Santeri Ruuskanen, Eero Saarro, Carola Maria Caivano et al.· bioRxiv· 0 citations
The importance of brain state in connectome-based cognitive prediction is highlighted and the brain cognition gap is introduced as a potentially informative, dopamine-modulated marker of vulnerability to compromise brain function.
M. Esmaeili, E. Bjørkeli, Robin Pedersen et al.· eLife· 0 citations
These findings support the contribution of Transformer processing to language-map prediction and distinguish it from subject-specific spatial conditioning and relative uncertainty estimation.
D. Di Giovanni, D. L. Collins· bioRxiv· 0 citations
A novel approach for deep extraction of causally informed features via restricted architecture (DECIFRA), designed for multivariate time series modeling that captures functionally grounded brain connectivity patterns, with certain dynamics associated with age or schizophrenia classification tasks and others consistentl...
P. Popov, Cristian Morasso, Giorgio Dolci et al.· IEEE Transactions on Medical...· 0 citations
A neuroinformatics framework for systematic evaluation of multiple FC measures under small-sample rs-fMRI conditions and explicitly addresses data leakage and overfitting through strict cross-validation and training-only feature selection is proposed.